Beyond Correlations and Averages

Individual Differences in Causal Effects of Religious Attendance

Joseph A. Bulbulia

Victoria University of Wellington, New Zealand

Thanks

Background

Overview

  1. New Zealand Attitudes and Values Study (NZAVS):

    • A nationally representative panel study, initiated in 2009.

    • Sample frame is drawn randomly from the NZ Electoral Roll.

    • Postal questionnaire (coverage; retention ~ 70-80%).

    • Large multidisciplinary research team (>60).

    • Focus on personality, social attitudes, values, religion, employment, prejudice \dots

    • Tracks nearly 1% of New Zealand’s adult population annually, with 76,409 participants to date.

  2. Religion’s Causal Effects on Cooperation

  3. Religion’s Causal Effects on Multi-Dimensional Flourishing

  4. Religion’s Causal Effects on Personality

What will we learn today?

Scope and Assumptions

  • Dis-engangling Causation from Correlation in Effects of Religious Service: Good-Life/ Cooperation
  • Focus on “Individual Differences”
  • Evolutionary Interests About How These Domains Are Related

Daily Counts of Processed NZAVS Surveys: Repeated Measures –> Causal Insights

Two Questions/ Two Domains

Eligibility

  • Responded to the time 10 (2018/19) New Zealand Attitudes Study
  • Information about religious service at baseline (yes or no).
  • Inverse Probability of Censoring Weights for Attrition
  • ‘Censored’ if responded during COVID lockdowns (NZAVS 10)
  • A total of 46,377 individuals met these criteria.

Method

  • Clearly stated causal estimands
  • Semi-parametric Machine Learning
  • Censoring models to recover what would have happened to population.
  • New Zealand Census Weights to for population inference
  • Cross-validation: Models always evaluated on unseen data

Context

Sample Demography

Table 1

Demographic statistics by religious denomination (NZAVS wave 10, approx~1.4% of NZ adult population).

Two Types of Esimands

Figure 1: ATE: Three-Wave: Binary Intervention

STUDY 1: Religious Service and The Good Life

1.1 Three Waves Good Life: ATE Causal Forests and Binary Exposure

Figure 2: ATE: Three-Wave: Binary Intervention

1.2 Good Life: Compare 3 waves Binary vs. 3 waves ALL/NONE

Figure 3: ATE: Three-wave Shift Intervention

1.3 Good Life Validation: Compare 3 waves Shift+1 Religious Service with 3 waves Shift+1 Socialising Hours

Figure 4: ATE: Three-wave Soft Intervention

1.4 Good Life: Compare 3 waves ALL/NONE vs 6 waves ALL/NONE

Figure 5: ATE: Six-wave Soft Intervention GAIN: +1

1.5 Good Life: Compare 6 waves ALL/NONE vs 6 waves SHIFT+1

Figure 6: ATE: Six-wave Soft Intervention LOSS: -1

1.6 Good Life: Compare 6 waves SHIFT+1 vs 6 waves Shift-1

Figure 7: CATE RATE

SUMMARY STUDY 1: GOOD LIFE

  • Depends on which effect (single intervention vs five interventions/e.g. shift up/down)
  • Unlikely to be due to hours of socialising (an overly conservative sensitivity analysis)
  • Signals are detectable in binary forests
Figure 8: CATE RATE

STUDY 2: Causal Effects of Religous Service on Cooperation

2.1 Three Waves Cooperation: ATE Causal Forests and Binary Exposure

Figure 9: ATE: Three-Wave: Binary Intervention

2.2 Three Waves Cooperation: Compare 3 waves Binary vs. 3 waves ALL/NONE

Figure 10: ATE: Three-wave Shift Intervention

2.3 Three Waves Cooperation: Compare 3 waves ALL/NONE vs. 3 waves SHIFT+1

Figure 11: ATE: Three-wave Soft Intervention

2.4 Six Waves Cooperation: Compare 3 waves ALL/NONE vs 6 waves ALL/NONE

Figure 12: ATE: Six-wave Soft Intervention GAIN: +1

2.5 Six Waves Cooperation: Compare 6 waves SHIFT+1 vs 6 waves Shift-1

Figure 13: ATE: Six-wave Soft Intervention LOSS: -1

Summary Study 2: Cooperation

STUDY 3: Individual Differences

3.1 Indv Diff: RS->Good Life: RATE

3.2 Indv Diff: RS->Good Life: UPLIFT

3.4 Indv Diff: RS->Good Life: POLICY 1

3.3 Three-wave Individual Differences: CATE Policy Tree: 1

3.4 Three-wave Individual Differences: CATE Policy Tree: 2

3.5 Three-wave Individual Differences: CATE Policy Tree: 3

3.6 Three-wave Individual Differences: CATE Policy Tree: 4

Overall Findings

Important Limitations

  • 76,409 individuals who have participated in the New Zealand Attitudes and Values Study since 2009.
  • Templeton Religion Trust: #0198 / #0418
  • University of Auckland / Victoria University of Wellington / Georgia State University
  • Graduate Students & Colleagues

References